ᴱ√TUAKK “*axe”
The root ᴱ√TUAKK appeared in the Qenya Lexicon of the 1910s as the basis for ᴱQ. tuakka “axe” (QL/95). The majority of axe words from this period seem to be derived from an unattested root *ᴱ√PELEK, including ᴱQ. pelekko/G. peleg “axe” and ᴱQ. pelekta-/G. pelectha- “hew” (QL/73; GL/64). ᴱ√PELEK thus formed a more comprehensive paradigm than ᴱ√TUAKK, and may have been an extension of ᴱ√PELE² “*revolve” referring to the pivoting motion of swinging an axe. Hints of ᴱ√PELEK can still be seen in ᴱN. peleg “axe” from Early Noldorin Word-lists from the 1920s (PE13/152). In later writings, Tolkien used S. hathol for “axe”. However, I think it is worth positing a Neo-Root ᴺ√PELEK to salvage the Early Qenya words.
ᴹ√TUB “to fall low[?], go down (below normal ground level), (esp.) to go down (sink, dive) into water”
Tolkien used several similar roots over his lifetime as the basis for Q. Utumno and S. Udûn, the underground stronghold of Melkor. The earliest of these was unglossed ᴱ√TUM(B)U in the Qenya Lexicon of the 1910s with derivatives ᴱQ. tumbo “dale, valley” and ᴱQ. tumna “deep, profound, dark or hidden” (QL/95). It also had derivatives in the contemporaneous Gnomish Lexicon like G. tûm “valley” and G. tumla- “excavate, hollow out” (GL/71-72).
The root ᴹ√TUB appeared unglossed in The Etymologies of the 1930s with derivatives like ᴹQ. tumbo/N. tum “deep valley” and ᴹQ. tumna/N. tofn “lowlying, deep, low”, as well as ᴹQ. Utumno/N. Udūn (Ety/TUB; EtyAC/TUB). The root reappeared in a rejected page of roots from the Quenya Verbal System of the 1940s where it had the verbal sense “to fall low[?], go down, below normal ground-level, esp. to go down (sink, dive) into water” (PE22/147). In this 1940s document the root had derivatives similar to those in The Etymologies, as well as a verb derivative ᴹQ. tumba- “to cast down (into a pit[?])”. One indication that this verbal sense was not a new idea was the verb ᴱQ. tum- “dive” from Early Qenya Word-lists of the 1920s.
Tolkien’s continued use of both Q. tumbo and S. tum for “valley” indicate the ongoing validity of ᴹ√TUB, but in drafts of The Silmarillion from the 1950s Tolkien derived Q. Utumno from ✶Utupnŭ and the root √TUI, probably a malformed √TUP (MR/69); see the entry on √TUP for a discussion of that root.
ᴱ√TUÐU “kindle”
The root ᴱ√TUŘU “kindle” [TUÐU] appeared in the Qenya Lexicon of the 1910s with variant forms ᴱ√TUSO and ᴱ√TUSU as well as derivatives like ᴱQ. tunda- “kindle”, ᴱQ. turu “wood, properly firewood”, and ᴱQ. tusturin “match” (QL/96). The contemporaneous Gnomish Lexicon had a similar primitive form tudh- with derivatives like G. tund “log for the fire”, G. tusta- “inflame, kindle, set light to, burn”, and G. tuthli “match” (GL/72). However, the Gnomish Lexicon also had words like G. drui “wood, forest” and G. duru “wood; a pole, beam, or log” (GL/31). Since initial d- > t- in Early Qenya, it seems likely that ᴱ√TUŘU “kindle” may also represent a blending with an unattested root *ᴱ√DURU “wood”.
Many years later Tolkien gave a hypothetical root √TUD in contrasted to √TUL to illustrate certain principles of etymological variations (VT48/25). It is not clear whether this √TUD is related to earlier ᴱ√TUÐU “kindle”. Nevertheless, I think it is worth positing a Neo-Root ᴺ√TUD “*firewood, kindling” to salvage Early Qenya and Gnomish words of similar meaning.
ᴹ√TUG “*strength”
An unglossed root in The Etymologies of the 1930s with derivatives like ᴹQ. tuo/N. tû “muscle, sinew; vigour, physical strength” and ᴹQ. tunga/N. tong “taut, tight (of strings, resonant)” (Ety/TUG). It was also the basis for the name N. Tuor “strength-vigor”, the only place where this name was given a clear etymology, though later Tolkien decided this name was actually from the language of the Edain (PM/348, 364 note #49).
ᴹ√TUK “draw, bring”
A root in The Etymologies of the 1930s glossed “draw, bring” with derivatives like ᴹQ. tuk- “draw” and N. tog- “{lead}, bring” (Ety/TUK; EtyAC/TUK). It might be a later iteration of unglossed ᴱ√TUKU from the Qenya Lexicon of the 1910s with derivatives like ᴱQ. tuku- “go in search of, look for, fetch” and G. tug- “hit, reach mark; light on, chance on, find; arrive; chance, occur; suit, fit, be convenient, ‘do’” (QL/95; GL/71). However, the 1910s and 1930s meanings of the root seem to be quite different.
ᴱ√TUKSA “*hundred”
An unglossed root in the Qenya Lexicon of the 1910s with variant ᴱ√TUKU⁽²⁾ and derivatives like ᴱQ. tuksa “gross, 144” and ᴱQ. tuksa pínea or tuksa lempea “100”, the variation in meaning likely due to the use of the word in duodecimal vs. decimal counting (QL/95). In the Early Qenya Grammar of the 1920s, ᴱQ. tuksa was used only with the sense “100” (PE14/49, 83). This is the last clear reference to an Elvish word for “100”, so I recommend keeping a Neo-Root ᴺ√TUKSA for this purposes, though Q. haranyë “century” (LotR/1108) hints that Tolkien may have used some of other construction for “100” in his later writings.
√TUL “come, approach, [ᴹ√] move towards (point of speaker); [ᴱ√] fetch, bear, bring; (originally) uphold, support, bear, carry”
The root √TUL was used for Elvish words having to do with motion towards a speaker for much of Tolkien’s life, but the precise meaning evolved over time. The earliest appearance of this root was as ᴱ√TULU “fetch, bear, bring; move, come”, but with an original sense = “uphold, support, bear, carry” (QL/95). It had a derived verb ᴱQ. tulu- matching the verbal sense of the root along with an added sense “produce, bear fruit”, but it had other derivatives like ᴱQ. tulma “bier, tray” and ᴱQ. tulwe “tall thin pillar, standard, pole; banner” connected to the older sense of the root. It had a similar divergence of meaning in its derivatives from the contemporaneous Gnomish Lexicon, such as G. taul “a pillar” vs. G. tul- “bring; come to” (GL/69, 71). This 1910s root ᴱ√TULU may also have been connected to ᴱ√TḶPḶ which likewise had derivatives having to do with “support” (QL/93).
In The Etymologies of the 1930s the root ᴹ√TUL had the gloss “come, approach, move towards (point of speaker)” with derivatives like ᴹQ. tul-/N. tol- “come” and ᴹQ. tulta-/N. toltha- “send for, fetch, summon” = “(orig.) make come” (Ety/TUL); in the 1930s the “support” words seem to have been transferred to (probably unrelated) ᴹ√TULUK. √TUL “come” was mentioned regularly in Tolkien’s later writings (PE22/103; PE17/188; PE22/156), in one 1969 note with the extra gloss “approach” (PE22/168), so the 1930s senses for the root seem to have been retained thereafter.
ᴹ√TULUK “*support, prop; steady, firm”
An unglossed root in The Etymologies of the 1930s with derivatives like ᴹQ. tulka “firm, strong, immoveable, steadfast”, N. tolog “stalwart, trusty”, and ᴹQ. tulko/N. tulu “support, prop” (Ety/TULUK). The “steady, firm” senses of the 1930s root are consistent with derivatives of the earlier root ᴱ√TULUK from the Qenya Lexicon of the 1910s with derivatives like ᴱQ. tulka- “fix, stick in, set up, establish”, G. tulga- “make firm, confirm, settle, steady; comfort” and ᴱQ. tulunka/G. tulug “steady, firm” (QL/95; GL/71); nearby ᴱ√TḶKḶ (with no derivatives) may also have been related (QL/93).
However, the Qenya Lexicon of the 1910s had a distinct root ᴱ√TḶPḶ with derivatives like ᴱQ. talpa/G. talp “support, prop”, ᴱQ. tulpu-/G. talpa- “prop up”, and G. clib- “uphold, support” (QL/93; GL/26, 69). This ᴱ√TḶPḶ may have been related to ᴱ√TULU as presented in Qenya Lexicon, which in this document had the more ancient sense “uphold, support” (QL/95). Thus I think the 1930s derivatives of ᴹ√TULUK having to do with “support” may have been transferred from 1910s ᴱ√TḶPḶ.
ᴹ√TUMPU “hump”
A root in The Etymologies of the 1930s glossed “hump”, with Quenya and Noldorin derivatives of the same meaning (Ety/TUMPU). It might be a later iteration of the root ᴱ√KUPU “hump” from the Qenya Lexicon of the 1910s which had derivatives like ᴱQ. kumpo “pile” and ᴱG. cub “hollow” (QL/49; GL/27). Alternately, it could be a later interation of ᴱ√TUMU “swell (with idea of hollowness)” (gloss marked with “?”) from the same document, but the one clear derivative of ᴱ√TUMU was ᴱQ. tumbe “trumpet, large horn” (QL/95), which seems unrelated to 1930s ᴹ√TUMPU.